Archives

Disabled Innovators & Innovations

Leave a comment

Disability culture, disability innovation, and DIY efforts have long been an important part of the work that disabled leaders and resesearch values, studies, and contributes to. Over the last year, we’ve been exploring this through a couple of perspectives:

First, we’ve been developing A11yhood.org, a website that automates the collection of open source accessibility solutions across a wide variety of media from 3D printing to knitting to software, and supports search and exploration. We searched for words like “accessibility” and found over 1440 repositories/things. This project was recently launched at the Open Source and Accessibility Summit and is supported by the GitHub Tides Foundation, NIDILRR’s RERC program (90REGE0026) and in collaboration with other open-source and accessibility focused organizations and leaders including GitHub’s Ed Summers, CAOS and GOAT.

Next, we have been talking to disabled innovators. We presented a paper at ASSETS 2025 that highlights cultural processes of finding community and building solidarity, valuing disabled agency and knowledge, and rejecting ableist norms. To see how these cultural aspects might inform accessibility technology design, we studied accessibility technologies made by disabled people for disabled people – interviewing disabled innovators who had created and disseminated accessibility technologies. We asked these innovators to share their stories and reflect on goals and values they imbued in their innovations. We analyzed how cultural themes of belonging, knowledge, and creativity influenced their work. Our work highlights the potential of a cultural lens in aligning accessibility technology with disabled people’s values as well as unearthing new directions for inquiry for the field: Exploring Disability Culture Through Accounts of Disabled Innovators of Accessibility Technology, by Aashaka Desai, Jennifer Mankoff, Richard E. Ladner
(ASSETS ’25)

MatPlotAlt

Leave a comment

MatplotAlt is an open-source Python package for easily adding alternative text to matplotlib figures. MatplotAlt equips Jupyter notebook authors to automatically generate and surface chart descriptions with a single line of code or command, and supports a range of options that allow users to customize the generation and display of captions based on their preferences and accessibility needs.

Our evaluation indicates that MatplotAlt’s heuristic and LLM-based methods to generate alt text can create accurate long-form descriptions of both simple univariate and complex Matplotlib figures. We find that state-of-the-art LLMs still struggle with factual errors when describing charts, and improve the accuracy of our descriptions by prompting GPT4-turbo with heuristic-based alt text or data tables parsed from the Matplotlib figure.

Here is some example ALT text generated for the pie chart shown below. A variety of examples can be found in the MatPlotAlt documentation.

A pie chart titled ’percentage of annual sunshine’. There are 12 slices: jan (3.19%), feb (4.993%), mar (8.229%), apr (9.57%), may (11.7%), june (12.39%), july (14.42%), aug (12.99%), sep (10.22%), oct (6.565%), nov (3.329%), and dec (2.404%). The data has a standard deviation of x=4.006, an average of x=8.333, a maximum value of x=14.42, and a minimum value of x=2.404. The data strictly increase up to their max at x=14.42, then strictly decrease.

A pie chart titled ’percentage of annual sunshine’. There are 12 slices: jan (3.19%), feb (4.993%), mar (8.229%), apr (9.57%), may (11.7%), june (12.39%), july (14.42%), aug (12.99%), sep (10.22%), oct (6.565%), nov (3.329%), and dec (2.404%). The data has a standard deviation of x=4.006, an average of x=8.333, a maximum value of x=14.42, and a minimum value of x=2.404. The data strictly increase up to their max at x=14.42, then strictly decrease.

Kai Nylund, Jennifer Mankoff, Venkatesh Potluri: MatplotAlt: A Python Library for Adding Alt Text to Matplotlib Figures in Computational Notebooks. Comput. Graph. Forum 44(3) (2025)